Acute myocardial infarction with ST-segment elevation (STEMI) remains a leading cause of morbidity and mortality worldwide. Although advances in reperfusion therapy have reduced early mortality, many patients later develop adverse ventricular remodeling (AVR), which increases the risk of heart failure and cardiovascular death. Current imaging methods, such as echocardiography and cardiac magnetic resonance (CMR), provide valuable prognostic information but have limitations in availability, cost, and their ability to predict AVR early and individually.
Spectral computed tomography (CT) is an emerging imaging technique that can characterize myocardial tissue, quantify infarct size, assess microvascular obstruction, and detect complications, with lower contrast and radiation requirements compared to conventional CT. In parallel, circulating microRNAs (miRNAs) have been identified as stable and non-invasive biomarkers that reflect key biological processes in post-infarction remodeling. Several miRNAs are linked to fibrosis, apoptosis, and ventricular remodeling, suggesting their potential to complement imaging findings in risk prediction.
This study proposes a multicenter, prospective cohort of patients with STEMI and reduced left ventricular function to evaluate whether combining spectral CT tissue characterization with serum miRNA profiling can improve early prediction of AVR. The main objective is to generate and validate a multiparametric prognostic model integrating imaging and molecular biomarkers to identify high-risk patients who may benefit from closer monitoring and tailored therapeutic strategies.
Adults ≥18 years.
First STEMI treated with primary PCI.
Left ventricular ejection fraction (LVEF) ≤45% during index hospitalization.
Ability to provide informed consent.
Contraindication to iodinated contrast media.
Chronic kidney disease with eGFR <30 mL/min/1.73 m².
Prior myocardial infarction or known cardiomyopathy.
Contraindication to CT or MRI imaging.
Parallel
Blood analysis, spectral CT scan and cardiac magnetic resonance imaging (CMR) study
Procedure/SurgeryAfter providing informed consent, patients will undergo blood sampling, a spectral CT scan scheduled between the 3rd and 7th day of hospitalization (acute phase), and a cardiac magnetic resonance imaging (CMR) study performed within a maximum of 72 hours from the CT.
Blood analysis and spectral CT study
Procedure/SurgeryPatients who meet the inclusion criteria for the control group will be invited to participate in the study. They will undergo blood sampling and the planned spectral CT study, including a late iodine enhancement acquisition.
Incidence of adverse ventricular remodeling at 6 months, defined as an increase in LV end-diastolic volume ≥20% compared with baseline.
The primary endpoint will be the occurrence of adverse ventricular remodeling (AVR) as assessed by cardiac magnetic resonance imaging (CMR). AVR will be defined as an increase in left ventricular end-diastolic volume (LVEDV) of ≥20% compared with baseline values obtained during the acute phase (3-7 days post-STEMI). In cases where CMR is not feasible, spectral CT-derived LV volumes will be used as a contingency reference, given their validated correlation with CMR measurements.
Infarct size characterization by spectral CT vs. CMR.
Quantitative assessment of infarct size will be performed using spectral CT (late iodine enhancement, iodine concentration maps, and non-contrast acquisitions for edema) and compared with CMR (late gadolinium enhancement, T1/T2 mapping, and extracellular volume fraction).
Myocardial characterization by spectral CT vs. CMR.
Quantitative assessment of myocardial markers such as microvascular obstruction (MVO), intramyocardial hemorrhage, and myocardial edema will be evaluated to determine the diagnostic accuracy and reproducibility of spectral CT (late iodine enhancement, iodine concentration maps, and non-contrast acquisitions for edema) and compared with the gold standard CMR (late gadolinium enhancement, T1/T2 mapping, and extracellular volume fraction).
Correlation of miRNA signatures with remodeling.
Serum miRNA profiles obtained at baseline (acute phase) and at 6 months will be analyzed to identify differential expression patterns associated with adverse remodeling. Specific miRNAs previously linked to fibrosis, apoptosis, and inflammation (e.g., miR-1, miR-21, miR-30a-5p, miR-133a, miR-210-3p, miR-221-3p) will be quantified by RT-qPCR. Qualitative correlation between miRNA expression levels and imaging-defined markers of remodeling will be tested, including potential sex-specific differences.
Predictive accuracy of integrated models (spectral CT + miRNA vs. conventional predictors)
Predictive models will be developed combining spectral CT-derived parameters, miRNA signatures, and conventional clinical predictors (age, sex, infarct location, Killip class, biomarkers). Model performance will be assessed using machine learning approaches (logistic regression, random forest, XGBoost), with internal validation through repeated 10-fold cross-validation. Predictive accuracy will be reported using receiver-operating characteristic (ROC) curves, area under the curve (AUC), calibration statistics, and Shapley values for interpretability. All parameters sill be used simultaneously to predict integrated models, therefore they cannot be sepparated in different outcomes.
Click on the lead sponsor to view all of their trials.
Instituto de Investigación Biomédica de Salamanca
Lead sponsor
Fundación Centro Nacional de Investigaciones Cardiovasculares Carlos III
Collaborator
